300 



PLANT-BREEDING 



increase the crown of pistillodous stamens, while adverse 

 influences reduce the change and thus favor the production 

 of the normal organs. 



The same influences, however, govern the degree of 

 development of the whole plant, the height and the thickness 

 of the stem, the size of the flower and the fruit, the number 

 and the strength of the branches, and even the extent of the 

 foliage. Hence we may expect a correlation between the 

 abnormality and the growth of all the normal parts, and it is 

 quite easy to control this. The best means is to choose the 

 ripe capsules. Their size and weight is evidently the result 

 of the activity of the whole plant, during the whole time 

 of its life, and thus may be considered as a standard for 

 comparison. 



In a bed of pistillodous poppies, we pick out a group 

 of fullgrown fruits and arrange them according to their 

 size, taking as a measure either the height or the weight of 

 the central capsule without the crown of secondary organs. 

 Then we compare the degree of development of the anomaly 

 with this size, and we find an almost complete parallelism. 

 The smallest fruits are devoid of appendixes or nearly so, 

 and the largest ones bear the richest crowns. Between 

 these extremes the number of the altered stamens regularly 

 increases with the size of the fruit, hardly any deviation 

 from the most complete parallelism occurring. 



By this investigation the normal and the abnormal 

 development are proven to be as closely correlated as might 

 be expected. So it is also in other instances. Ordinarily, 

 of course, abnormalities are too rare to permit of such a 

 complete comparison, but notwithstanding this, common 

 experience shows them to be connected with strong growth 

 and favorable life conditions. This relation is often so 

 striking that in former times it was simply assumed that 

 monstrosities could be produced by an excess of nourishment, 



